GPS: A Policy-driven Sampling Approach for Graph Representation Learning

TL;DR

The paper introduces GPS, an adaptive, policy-driven sampling method for graph representation learning that improves training efficiency and accuracy on large-scale graphs by guiding neighbor selection through an adaptive policy.

Contribution

It proposes a novel adaptive sampling strategy using a policy algorithm to enhance graph embedding learning, outperforming existing methods on multiple benchmarks.

Findings

Outperforms baseline methods by 3%-8% on graph classification benchmarks.

Achieves state-of-the-art results on real-world large-scale graph datasets.

Enhances message aggregation and node embedding quality through adaptive neighbor selection.

Abstract

Graph representation learning has drawn increasing attention in recent years, especially for learning the low dimensional embedding at both node and graph level for classification and recommendations tasks. To enable learning the representation on the large-scale graph data in the real world, numerous research has focused on developing different sampling strategies to facilitate the training process. Herein, we propose an adaptive Graph Policy-driven Sampling model (GPS), where the influence of each node in the local neighborhood is realized through the adaptive correlation calculation. Specifically, the selections of the neighbors are guided by an adaptive policy algorithm, contributing directly to the message aggregation, node embedding updating, and graph level readout steps. We then conduct comprehensive experiments against baseline methods on graph classification tasks from various perspectives. Our proposed model outperforms the existing ones by 3%-8% on several vital benchmarks, achieving state-of-the-art performance in real-world datasets.